Small Reservoirs Extraction in Semiarid Regions Using Multitemporal Synthetic Aperture Radar Images

In this paper, we introduce a novel framework for small reservoirs extraction in semiarid environment. The task is accomplished through the introduction of a pseudoprobability index derived from multitemporal synthetic aperture radar RGB images. These products are characterized by the ease of interpretation for nonexpert users, and the possibility to be processed using simple algorithms, allowing, in this case, for the definition of an ad hoc band ratio for feature extraction. The reliability of the proposed approach is demonstrated through a case study in Burkina Faso in which 19 reservoirs up to about 6000 m
$^2$
extent were tested. The obtained accuracy with respect to the available ground truth is higher than 88%.